Vascular proliferation is a pathologic hallmark of glioblastoma and studies have shown that the expression of pro-angiogenic factors such as vascular endothelial growth factor (VEGF) is one mechanism by which tumors induce the formation of new blood vessels. Antiangiogenic therapy has been shown to rapidly decrease vascular permeability, reduce cerebral edema, and prolong progression free survival (PFS) in some patients with recurrent glioblastoma. However, patients with intrinsic resistance do not benefit from anti-VEGF therapy and those that initially respond acquire resistance limiting the long-term effectiveness of this approach. The mechanisms of resistance to anti-VEGF therapy are potentially many but recent reports suggest that bone marrow derived cells (BMDCs) may be important mediators of resistance. We have developed a novel clinical trial designed to potentially modulate the development of glioblastoma resistance to bevacizumab. Patients with recurrent glioblastoma will be randomized to receive standard dose bevacizumab or low-dose bevacizumab combined with lomustine given on day 3 during the "normalization window" to maximize drug delivery to tumor. The lower dose of bevacizumab will be utilized to both improve drug delivery and to prevent or delay the development of severe hypoxia which we have shown to be a prominent feature of glioblastoma tumors treated with chronic anti-VEGF therapy. To evaluate potential modulators of glioblastoma sensitivity and resistance to anti-VEGF therapy, this trial will assess changes in circulating chemokines and myeloid cells in the standard dose versus low dose arms. In Aim 1, we will determine if baseline plasma levels of monocytic and granulocytic chemokines are predictive of response to anti-VEGF therapy. Aim 2 will assess the recruitment of myeloid cells following anti-VEGF therapy. Aim 3 will assess the association between plasma chemokines, the number of myeloid cells infiltrated into glioblastoma samples and histologic markers of vascular proliferation at the time of surgery for recurrent glioblastoma. These studies are critical to establish which biomarkers are useful in predicting response and monitoring treatment progress, as well as for forming a baseline for future trials that incorporate therapies targeting myeloid cell populations. Inclusion of biomarker measurements into this phase II study represent unique strengths of our institution and will enhance our ability to assess the potential efficacy of an alternative bevacizumab dosing schedule in a multidimensional manner. PUBLIC HEALTH RELEVANCE: Despite the promising radiographic responses seen in patients with recurrent glioblastoma treated with antiangiogenic therapy, the long term effectiveness of this therapy remains limited by the development of resistance. Our clinical trial will attempt to modulate and delay the development of resistance to anti-VEGF therapy while exploring plasma chemokines and the myeloid cells they recruit as predictive markers of sensitivity and resistance to therapy.